Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsBlog/PodcastSlackPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlog/PodcastGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 5-RES-JB
    6. 10759
    Prev
    Next
    INDEX
    Explanations

    special characters and specific phrases used for formatting or identification, such as a unique symbol followed by a date

    oai_token-act-pair · gpt-3.5-turbo

    the presence of the character "âĢĭ" in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 5-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.hook_resid_pre
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    24,576
    Data Type
    torch.float32
    Hook Point
    blocks.5.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    5
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    oresc
    -0.88
    orescence
    -0.84
    pless
    -0.76
    ffield
    -0.72
    creen
    -0.71
     rooting
    -0.70
    otine
    -0.67
    glers
    -0.65
    arsity
    -0.64
    obbies
    -0.64
    POSITIVE LOGITS
      
    1.02
      
    0.96
    Leaks
    0.90
    âĢ¢âĢ¢
    0.85
    Mich
    0.79
     
    0.77
    Bloomberg
    0.77
    Kings
    0.77
    Harris
    0.75
    Toronto
    0.73
    Activations Density 0.006%

    No Known Activations